Remeasurements of weathering rates at six sites on the base of the balustrade of Saint Paul's Cathedral gave substantially lower mean results (0. 06 mm a^-1, 1980-1985) than for earlier results (0.14 mm a^-1, 1980-1981). Statistical analysis of the data showed that for five of the sites there were no significant differences between the measurement periods. The only significant spatial difference was for the southwest (exposed) site which was greater than the other sites for 1980-1981, but not at later times. This site also showed the only significant decrease over time. If this site is included in the statistical analysis, the whole set shows a significant decrease over time; omitting this site from the data removes any significant differences over time. The results for this one site for 1980-1981 thus have significant bias on the data body as a whole. Methodological problems, such as rock erosion by the probe and temperature effects, are thought to have minimal effect on the interpretation of the data. Calibrations showed that measurements were accurate to ±0.006 mm and that only results of a change greater than or equal to 0.02 mm were reliably interpretable as different. Remeasurements of the Lead Plug Index (LPI) for the top of the balustrade in 1987 gave a mean rate for 1718-1985 of 0.081 mm a^-1. This is not significantly different from the 1718-1980 mean rate of 0.078 mm a^-1 but the data are not directly comparable, the 1987 sample being smaller than the 1718-1980 set. For the 1987 data, the only statistically significant differences are the southeast rate much greater than northeast and for a rise in rate in the northeast quadrant. The LPI data are comparable to most 1980-1985 MEM data. The conclusion is that most erosion rates have neither increased nor decreased significantly in the measurement time, except for a significant decrease at one (southwest) site from 1980-1981 to 1981-1982. Such a sustained level of erosion gives cause for concern in the context of stone conservation. Causative factors of air quality are discussed and while sulfur dioxide levels have decreased, nitrous dioxide and smoke levels show increases. An overall decrease in stone decay in association with the decrease in sulfur dioxide levels is thus not in evidence.